An accurate and complete medication list on a patient's electronic health record (EHR) is critical to prevent prescribing and administration errors. Stage 1 of Meaningful Use requires certified EHRs to be capable of providing a user with the ability to perform medication reconciliation. However, most previous studies have taken place in the inpatient setting, while medication reconciliation in the outpatient setting is importnt and challenging. In addition, clinical notes contain critical medication information that also need to be reconciled. Our goal of this study is to develop novel methods and a system using natural language processing (NLP) and other technologies to facilitate the medication reconciliation process in the ambulatory setting. Our specific aims are to : 1) identify the requirements, use cases, work flow issues, barriers to and facilitators of using clinical notes and a NLP-based system in the medication reconciliation process; 2) design a generic system architecture and an application that integrates an NLP system and a web-based user interface within an existing medication reconciliation system; 3) pilot this study in two primary care clinics and measure the utilization, usability, performance and feasibility of the proposed methods and the tool; and 4) distribute our methods and the tool and to make them widely available to other researchers and healthcare institutions for non-commercial use.